| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.12, | |
| "eval_steps": 4, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 102.28898620605469, | |
| "learning_rate": 2e-05, | |
| "loss": 6.6367, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 7.300913333892822, | |
| "eval_runtime": 1.3523, | |
| "eval_samples_per_second": 8.873, | |
| "eval_steps_per_second": 4.437, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 103.4541015625, | |
| "learning_rate": 4e-05, | |
| "loss": 7.0616, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 67.47515869140625, | |
| "learning_rate": 6e-05, | |
| "loss": 4.686, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 72.36919403076172, | |
| "learning_rate": 8e-05, | |
| "loss": 2.3866, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.7137572169303894, | |
| "eval_runtime": 1.3532, | |
| "eval_samples_per_second": 8.868, | |
| "eval_steps_per_second": 4.434, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 16.83085060119629, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6844, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 25.897714614868164, | |
| "learning_rate": 0.00012, | |
| "loss": 0.914, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 18.89151382446289, | |
| "learning_rate": 0.00014, | |
| "loss": 0.63, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 27.15555763244629, | |
| "learning_rate": 0.00016, | |
| "loss": 0.948, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 1.0445994138717651, | |
| "eval_runtime": 1.356, | |
| "eval_samples_per_second": 8.85, | |
| "eval_steps_per_second": 4.425, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 20.812381744384766, | |
| "learning_rate": 0.00018, | |
| "loss": 1.0285, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 56.3886604309082, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3756, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 6.24803352355957, | |
| "learning_rate": 0.00019981755542233177, | |
| "loss": 0.5178, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 8.379430770874023, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 0.6822, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.3959709405899048, | |
| "eval_runtime": 1.3583, | |
| "eval_samples_per_second": 8.835, | |
| "eval_steps_per_second": 4.417, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 20.744348526000977, | |
| "learning_rate": 0.00019836199069471437, | |
| "loss": 1.3762, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 4.800480842590332, | |
| "learning_rate": 0.0001970941817426052, | |
| "loss": 0.5248, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 11.284302711486816, | |
| "learning_rate": 0.00019547208665085457, | |
| "loss": 0.8094, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 5.787976264953613, | |
| "learning_rate": 0.0001935016242685415, | |
| "loss": 0.5222, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.9023411870002747, | |
| "eval_runtime": 1.3623, | |
| "eval_samples_per_second": 8.808, | |
| "eval_steps_per_second": 4.404, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 21.48629379272461, | |
| "learning_rate": 0.00019118998459920902, | |
| "loss": 0.8027, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 38.0982666015625, | |
| "learning_rate": 0.000188545602565321, | |
| "loss": 1.7772, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 10.824837684631348, | |
| "learning_rate": 0.00018557812723014476, | |
| "loss": 0.7737, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 9.1353120803833, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 0.534, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.4847445785999298, | |
| "eval_runtime": 1.3637, | |
| "eval_samples_per_second": 8.799, | |
| "eval_steps_per_second": 4.4, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 3.8411033153533936, | |
| "learning_rate": 0.00017871834806090501, | |
| "loss": 0.3201, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 23.888507843017578, | |
| "learning_rate": 0.00017485107481711012, | |
| "loss": 2.2541, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 8.5956392288208, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.8177, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 3.825141191482544, | |
| "learning_rate": 0.00016631226582407952, | |
| "loss": 0.4624, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.5740255117416382, | |
| "eval_runtime": 1.3655, | |
| "eval_samples_per_second": 8.788, | |
| "eval_steps_per_second": 4.394, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.558993101119995, | |
| "learning_rate": 0.00016167188726285434, | |
| "loss": 0.3714, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 11.759211540222168, | |
| "learning_rate": 0.00015680647467311557, | |
| "loss": 0.6562, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 96.2179183959961, | |
| "learning_rate": 0.00015173378141776568, | |
| "loss": 1.5141, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 31.022045135498047, | |
| "learning_rate": 0.00014647231720437686, | |
| "loss": 0.7753, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.3771994113922119, | |
| "eval_runtime": 1.3676, | |
| "eval_samples_per_second": 8.775, | |
| "eval_steps_per_second": 4.387, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 3.5004501342773438, | |
| "learning_rate": 0.0001410412805452757, | |
| "loss": 0.2649, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 5.16464376449585, | |
| "learning_rate": 0.00013546048870425356, | |
| "loss": 0.171, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 25.634010314941406, | |
| "learning_rate": 0.00012975030538552032, | |
| "loss": 0.9172, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 7.102908134460449, | |
| "learning_rate": 0.0001239315664287558, | |
| "loss": 0.3324, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 0.29374203085899353, | |
| "eval_runtime": 1.3678, | |
| "eval_samples_per_second": 8.773, | |
| "eval_steps_per_second": 4.387, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 6.236325263977051, | |
| "learning_rate": 0.0001180255037813906, | |
| "loss": 0.4932, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 4.445058345794678, | |
| "learning_rate": 0.0001120536680255323, | |
| "loss": 0.1284, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 6.94170618057251, | |
| "learning_rate": 0.00010603784974222861, | |
| "loss": 0.1547, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 5.656033039093018, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1973, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 0.5674905180931091, | |
| "eval_runtime": 1.3681, | |
| "eval_samples_per_second": 8.771, | |
| "eval_steps_per_second": 4.386, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 18.19667625427246, | |
| "learning_rate": 9.396215025777139e-05, | |
| "loss": 0.4884, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 17.964893341064453, | |
| "learning_rate": 8.79463319744677e-05, | |
| "loss": 0.5526, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 5.015590190887451, | |
| "learning_rate": 8.197449621860943e-05, | |
| "loss": 0.2116, | |
| "step": 39 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 62, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 13, | |
| "total_flos": 3414704284237824.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |